CN114595480A - Real-time passenger and driver matching method with personalized location privacy protection - Google Patents

Real-time passenger and driver matching method with personalized location privacy protection Download PDF

Info

Publication number
CN114595480A
CN114595480A CN202210212477.0A CN202210212477A CN114595480A CN 114595480 A CN114595480 A CN 114595480A CN 202210212477 A CN202210212477 A CN 202210212477A CN 114595480 A CN114595480 A CN 114595480A
Authority
CN
China
Prior art keywords
passenger
driver
user
node
layer
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202210212477.0A
Other languages
Chinese (zh)
Other versions
CN114595480B (en
Inventor
李向阳
吕超杰
张兰
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
University of Science and Technology of China USTC
Original Assignee
University of Science and Technology of China USTC
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by University of Science and Technology of China USTC filed Critical University of Science and Technology of China USTC
Priority to CN202210212477.0A priority Critical patent/CN114595480B/en
Publication of CN114595480A publication Critical patent/CN114595480A/en
Application granted granted Critical
Publication of CN114595480B publication Critical patent/CN114595480B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
    • G06F21/60—Protecting data
    • G06F21/62—Protecting access to data via a platform, e.g. using keys or access control rules
    • G06F21/6218—Protecting access to data via a platform, e.g. using keys or access control rules to a system of files or objects, e.g. local or distributed file system or database
    • G06F21/6245—Protecting personal data, e.g. for financial or medical purposes
    • G06F21/6254—Protecting personal data, e.g. for financial or medical purposes by anonymising data, e.g. decorrelating personal data from the owner's identification
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/901—Indexing; Data structures therefor; Storage structures
    • G06F16/9027—Trees
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/95—Retrieval from the web
    • G06F16/951—Indexing; Web crawling techniques
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/95—Retrieval from the web
    • G06F16/953—Querying, e.g. by the use of web search engines
    • G06F16/9537—Spatial or temporal dependent retrieval, e.g. spatiotemporal queries
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/06—Buying, selling or leasing transactions
    • G06Q30/0601—Electronic shopping [e-shopping]
    • G06Q30/0639—Item locations
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/40—Business processes related to the transportation industry

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Business, Economics & Management (AREA)
  • General Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • General Health & Medical Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Strategic Management (AREA)
  • Finance (AREA)
  • Marketing (AREA)
  • Bioethics (AREA)
  • Economics (AREA)
  • Accounting & Taxation (AREA)
  • General Business, Economics & Management (AREA)
  • Software Systems (AREA)
  • Computer Security & Cryptography (AREA)
  • Computer Hardware Design (AREA)
  • Development Economics (AREA)
  • Medical Informatics (AREA)
  • Human Resources & Organizations (AREA)
  • Primary Health Care (AREA)
  • Tourism & Hospitality (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention discloses a real-time passenger and driver matching method with personalized position privacy protection, which is used in a taxi taking system with a server side in communication connection with a plurality of user sides and comprises the following steps: step 1, establishing a model of a real-time matching area: the service end models a two-dimensional plane matrix region R for real-time matching of passengers and drivers into a four-branch tree T with the tree height of H; the server side issues the two-dimensional plane matrix region R and the quadtree T corresponding to the two-dimensional plane matrix region R to all the user sides, and the user sides comprise: passenger and driver clients; step 2, local personalized noise adding and submitting of the user position; step 3, matching passengers and drivers in real time; and 4, establishing communication between the passenger and the driver, namely completing matching. The method provides efficient and effective matching for real-time passengers and drivers on the server side through personalized position privacy protection and two privacy protection of different degrees, so that the position information is protected, and the matching accuracy is provided.

Description

Real-time passenger and driver matching method with personalized location privacy protection
Technical Field
The invention relates to the field of safety privacy, in particular to a real-time passenger and driver matching method with personalized position privacy protection.
Background
With the increase of data security requirements, data managers and operators have clear data protection responsibilities. It is a very difficult point to provide a personalized mechanism for protecting the positions of both parties between the passenger and the driver (hereinafter, the passenger and the driver are collectively referred to as the user) and a matching mechanism with certain accuracy. This is not only because of the need for different privacy protection for passengers and drivers, but also the need to provide a mechanism for matching multiple passengers and multiple drivers on a flat surface, reduce the matching distance between passengers and drivers, save time and resources, and improve efficiency.
Currently, a series of works exist for a user matching algorithm aiming at privacy protection: 1) non-private matching algorithms include bipartite graph matching algorithms; if the user can accept a certain delay, the matching distance between the real-time users is reduced greatly after the existing users are accumulated; 2) a privacy matching algorithm is adopted, the disturbance position of a driver is assumed to be given in advance, so that a matching model is established in advance, and passengers are matched one by one; 3) the privacy matching algorithm is established on the HST tree, and points (non-privacy points or privacy points) on a two-dimensional plane are changed into discrete points on the HST tree so as to realize quick matching. However, these methods are difficult to be directly applied to real-time user matching with personalized privacy protection because: 1) the existing non-privacy algorithms require the user to submit location data to the platform, i.e. the platform can directly access the location data of all passengers and drivers, whereas in private user matching, the data is not directly accessible by third parties other than the user; 2) the existing method needs to know data of one party in advance to establish a model in advance, and the data of one party cannot be known in advance in practice; 3) the existing method can not provide personalized privacy protection for the user, so that the user with low privacy requirement also needs to use higher privacy protection, and the usability of the position is reduced; users want different levels of location ambiguity and different degrees of privacy protection, and existing algorithms mainly take the location ambiguity or privacy protection into account, but do not consider both simultaneously.
In view of the above, the present invention is particularly proposed.
Disclosure of Invention
The invention aims to provide a real-time passenger and driver matching method with personalized position privacy protection, which can not only protect the position privacy of a user, but also accurately match passengers and drivers, thereby solving the technical problems in the prior art.
The purpose of the invention is realized by the following technical scheme:
the embodiment of the invention provides a real-time passenger and driver matching method with personalized position privacy protection, which is used in a taxi taking system with a server side in communication connection with a plurality of user sides and comprises the following steps:
step 1, establishing a model of a real-time matching area: the service end models a two-dimensional plane matrix region R for real-time matching of passengers and drivers into a four-branch tree T with the tree height of H; 1 node on the 0 th layer of the quadtree T is a root node, and the root node represents a region R; the layer 1 is provided with 4 nodes, and each node represents a horizontal, vertical and uniform four equal areas of the area R; the h-th layer has 4hEach node represents the horizontal, vertical and uniform four equal areas of the corresponding area of each node of the h-1 layer; the H-th layer has 4HEach node represents the horizontal, vertical and uniform four equal areas of the corresponding area of each node of the H-1 layer; numbering each node of the four-way tree T from top to bottom and from left to right from 0 in sequence, wherein the leftmost node of the h-th layer is numbered as (4)h-1) ÷ 3, the node number on the far right being (4)h+1-1)÷3-1；
The server side issues the two-dimensional plane matrix region R and the quadtree T corresponding to the two-dimensional plane matrix region R to all user sides, and the user sides include: passenger and driver clients;
step 2, local personalized noise adding and submitting of user positions: after the user side obtains the two-dimensional plane matrix area R and the quartering tree T corresponding to the two-dimensional plane matrix area R, a private one-hot vector x is obtained according to the personalized privacy parameters (e, h) of the user sidehUsing one-hot vector xhGeneration of perturbed one-hot vectors by truncation of geometric mechanisms
Figure BDA0003532769940000021
Make the perturbed one-hot vector
Figure BDA0003532769940000022
The privacy protection of the E-geo-indexing uishability of the line is met, and the user type and the disturbance vector are transmitted to the user side
Figure BDA0003532769940000023
Submitting the personalized privacy parameters (e, h) to the server;
in the personalized privacy parameters (epsilon, h), epsilon represents the personalized privacy budget parameters of the user terminal, and the value range is [0, + ∞ ]; h represents the height of the user selected to be published on the quadtree T through a user side, and the selection range is {0, 1, …, H };
the user types include: passengers, drivers;
step 3, matching passengers and drivers in real time: the server receives the user type and the disturbance vector sent by the user side in real time
Figure BDA0003532769940000024
And personalized privacy parameters (e, h) and continuously maintaining passenger queues LpAnd driver queue Ld(ii) a Then, the passenger and the driver are matched according to the existing user set and the new user, and the matching steps are as follows:
(31) if the type of the existing user only has passengers and the type of the newly arrived user is also the passenger, the newly arrived user is added into a passenger queue Lp；
(32) If the type of the existing user is only the companyIf the new user type is also the driver, adding the new user into the driver queue Ld；
(33) If the type of the existing user only has passengers and the type of the new user is a driver, the passenger queue L is arrangedpAll the passengers and the drivers are matched, and the passenger closest to the driver is selected to be matched with the driver;
(34) if the existing user type only has drivers and the new user type is passengers, the drivers are queued to LdAll drivers and passengers are matched, and the driver closest to the passenger is selected to be matched with the passenger;
and 4, establishing communication between the passenger and the driver: and after the server side is matched with a pair of passenger and driver, the respective IDs of the matched passenger and driver are sent to the other side, so that the matched passenger and driver establish communication, namely the matching is completed.
Compared with the prior art, the real-time passenger and driver matching method with personalized position privacy protection, provided by the invention, has the beneficial effects that:
because the quadtree is used as a model of the real-time matching area and a Truncation Geometry Mechanism (TGM) is matched, two privacy protections of position ambiguity at different levels and differential privacy at different degrees become possible; meanwhile, the distance and the matching mode between different layers are adopted, and the communication between the passenger and the driver after the matching is established, so that the passenger user and the driver user can be efficiently and effectively matched.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on the drawings without creative efforts.
FIG. 1 is a schematic flow chart of a real-time passenger and driver matching method with personalized location privacy protection according to an embodiment of the present invention;
FIG. 2 is a detailed flow chart of a real-time passenger and driver matching method provided by an embodiment of the invention;
FIG. 3 is a schematic diagram of a quadtree of a real-time passenger and driver matching method provided by an embodiment of the invention;
fig. 4 is a schematic diagram of the area where the quartering tree of the real-time passenger and driver matching method according to the embodiment of the present invention is located.
Detailed Description
The technical scheme in the embodiment of the invention is clearly and completely described below by combining the specific content of the invention; it is to be understood that the described embodiments are merely exemplary of the invention, and are not intended to limit the invention to the particular forms disclosed. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the present invention without making any creative effort, shall fall within the protection scope of the present invention.
The terms that may be used herein are first described as follows:
the term "and/or" means that either or both can be achieved, for example, X and/or Y means that both cases include "X" or "Y" as well as three cases including "X and Y".
The terms "comprising," "including," "containing," "having," or other similar terms of meaning should be construed as non-exclusive inclusions. For example: including a feature (e.g., material, component, ingredient, carrier, formulation, material, dimension, part, component, mechanism, device, process, procedure, method, reaction condition, processing condition, parameter, algorithm, signal, data, product, or article of manufacture), is to be construed as including not only the particular feature explicitly listed but also other features not explicitly listed as such which are known in the art.
The term "consisting of … …" is meant to exclude any technical feature elements not explicitly listed. If used in a claim, the term shall render the claim closed except for the inclusion of the technical features that are expressly listed except for the conventional impurities associated therewith. If the term occurs in only one clause of the claims, it is defined only to the elements explicitly recited in that clause, and elements recited in other clauses are not excluded from the overall claims.
Unless expressly stated or limited otherwise, the terms "mounted," "connected," and "secured," etc., are to be construed broadly, as for example: can be fixedly connected, can also be detachably connected or integrally connected; can be mechanically or electrically connected; they may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. The specific meaning of the above terms herein can be understood by those of ordinary skill in the art as appropriate.
When concentrations, temperatures, pressures, dimensions, or other parameters are expressed as ranges of values, the ranges are to be understood as specifically disclosing all ranges formed from any pair of upper, lower, and preferred values within the range, regardless of whether ranges are explicitly recited; for example, if a numerical range of "2 ~ 8" is recited, then the numerical range should be interpreted to include ranges of "2 ~ 7", "2 ~ 6", "5 ~ 7", "3 ~ 4 and 6 ~ 7", "3 ~ 5 and 7", "2 and 5 ~ 7", and the like. Unless otherwise indicated, the numerical ranges recited herein include both the endpoints thereof and all integers and fractions within the numerical range.
The terms "central," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," "counterclockwise," and the like are used in an orientation or positional relationship that is indicated based on the orientation or positional relationship shown in the drawings for ease of description and simplicity of description only, and are not intended to imply or imply that the referenced device or element must have a particular orientation, be constructed and operated in a particular orientation, and are therefore not to be considered limiting herein.
The following describes in detail the real-time passenger and driver matching method with personalized location privacy protection provided by the present invention. Details which are not described in detail in the embodiments of the invention belong to the prior art which is known to the person skilled in the art. Those not specifically mentioned in the examples of the present invention were carried out according to the conventional conditions in the art or conditions suggested by the manufacturer. The reagents or instruments used in the examples of the present invention are not specified by manufacturers, and are all conventional products available by commercial purchase.
As shown in fig. 1 and 2, an embodiment of the present invention provides a real-time passenger and driver matching method with personalized location privacy protection, which is used in a taxi taking system with a server communicatively connected to a plurality of clients, and includes:
step 1, establishing a model of a real-time matching area: the server side models a two-dimensional plane matrix region R for real-time passenger and driver matching into a four-branch tree T with the tree height of H (see figures 3 and 4); 1 node on the 0 th layer of the quadtree T is a root node, and the root node represents a region R; the layer 1 is provided with 4 nodes, and each node represents a horizontal, vertical and uniform four equal areas of the area R; the h-th layer has 4hEach node represents the horizontal, vertical and uniform four equal areas of the corresponding area of each node of the h-1 layer; the H-th layer has 4HEach node represents the horizontal, vertical and uniform four equal areas of the corresponding area of each node of the H-1 layer; numbering each node of the four-way tree T from top to bottom and from left to right from 0 in sequence, wherein the leftmost node of the h-th layer is numbered as (4)h-1) ÷ 3, the node number on the far right being (4)h+1-1)÷3-1；
The server side issues the two-dimensional plane matrix region R and the quadtree T corresponding to the two-dimensional plane matrix region R to all user sides, and the user sides include: passenger and driver clients;
step 2, local personalized noise adding and submitting of user positions: after the user side obtains the two-dimensional plane matrix area R and the quartering tree T corresponding to the two-dimensional plane matrix area R, a private one-hot vector x is obtained according to the personalized privacy parameters (e, h) of the user sidehUsing one-hot vector xhBy cutting off the tableWhich mechanism generates perturbed one-hot vectors
Figure BDA0003532769940000051
Let the one-hot vector
Figure BDA0003532769940000052
The privacy protection of the E-geo-indexing uishability of the line is met, and the user type and the disturbance vector are transmitted to the user side
Figure BDA0003532769940000053
Submitting the personalized privacy parameters (e, h) to the server;
in the personalized privacy parameters (epsilon, h), epsilon represents the personalized privacy budget parameters of the user terminal, and the value range is [0, + ∞ ]; h represents the height of the user selected to be published on the quadtree T through a user side, and the selection range is {0, 1, …, H };
the user types include: passengers, drivers;
step 3, matching passengers and drivers in real time: the server receives the user type and the disturbance vector sent by the user side in real time
Figure BDA0003532769940000054
And personalized privacy parameters (e, h) and continuously maintaining passenger queues LpAnd driver queue Ld(ii) a Then, the passenger and the driver are matched according to the existing user set and the new user, and the matching steps are as follows:
(31) if the type of the existing user only has passengers and the type of the newly arrived user is also the passenger, adding the newly arrived user into the passenger queue Lp；
(32) If the existing user type is driver only and the new user type is also driver, adding the new user into the driver queue Ld；
(33) If the type of the existing user only has passengers and the type of the new user is a driver, the passenger queue L is arrangedpAll the passengers and the drivers are matched, and the passenger closest to the driver is selected to be matched with the driver;
(34) if the existing user type only has drivers and the new user type is passengers, the drivers are queued to LdAll drivers and passengers are matched, and the driver closest to the passenger is selected to be matched with the passenger;
and 4, establishing communication between the passenger and the driver: after the service end is matched with a pair of passenger and driver, the respective IDs of the matched passenger and driver are sent to the opposite side, so that the matched passenger and driver establish communication (even if the matched passenger user side and driver user side establish communication), namely the matching is completed.
In step 2 of the method, the user side obtains a private one-hot vector x according to the personalized privacy parameters (e, h) of the user side in the following mannerhUsing one-hot vector xhGeneration of perturbed one-hot vectors by truncation of geometric mechanisms
Figure BDA0003532769940000061
Make the perturbed one-hot vector
Figure BDA0003532769940000062
Obtaining a disturbance vector according to the E-geo-indexing uishability privacy protection processing of the line where the line is positioned
Figure BDA0003532769940000063
The user side sends the user type and the disturbance vector
Figure BDA0003532769940000064
Submitting personalized privacy parameters (e, h) to the server, wherein the personalized privacy parameters comprise:
step 21, the user side obtains a private one-hot vector x according to the personalized parametersh: the user side determines a node n on the quartering tree T corresponding to the area where the user position is located by using a parameter h of the personalized privacy parameter (epsilon, h)h，iAnd connecting the node nh，iIs taken to be 1, the node nh，iThe value of the same-layer node is taken as 0, and the node n is usedh，iAll the nodes of the layer form a one-hot vector x by the values from left to righthThe node nh，iIn the one-hot vector xhAt position i;
step 22, generating a disturbed one-hot vector by using a truncation geometry mechanism
Figure BDA0003532769940000065
Let the one-hot vector
Figure BDA0003532769940000066
Processing satisfying the privacy protection of the e-geo-identifying uishability: one-hot vector xhCorresponding to the number of the nodes on the four-division tree T one by one according to a formula
Figure BDA0003532769940000067
Disturbing the two numbers to ensure that the disturbed same number x' meets the privacy protection mechanism M belonging to the E-geo-indentifying uishablility, wherein x in the formula1、x2The numbers of any two nodes on the quadtree T are shown, and x' is the same number formed by disturbance; pr [ M (x)1)＝x′]The representation mechanism M is x at the input1The probability that the time output is x'; d (x, x ') represents the difference between the absolute values of the nodes numbered x, x', i.e., d (x, x ') ═ x-x' |;
disturbing nodes with the number x on the h height in the quadtree T into the number x' according to the following formula, and performing truncation geometric mechanism processing to meet the requirement of E-geo-individissability, wherein the formula is as follows:
Figure BDA0003532769940000068
the meaning of each parameter of the above formula is: x, x' represent the node numbers,
Figure BDA0003532769940000071
Pr[M(x1)＝x′]the representation mechanism M is x at the input1The probability that the time output is x'; e is a natural constant; h represents the height of the node number x on the quadtree T;
converting the obtained disturbance number x' into corresponding disturbanceMotion vector
Figure BDA00035327699400000716
The formula satisfying the mechanism M is derived as follows, since the standard form e-DP TGM (trained Geometric mechanism) needs to add the sensitivity, and the sensitivity in TGM in the form e-geo-individinguishability is set to 1, that is, the mechanism M is satisfied; the derivation process comprises two parts: 1) the sum of the probabilities of obtaining all possible outputs from any input x via the mechanism M is equal to 1, i.e.
Figure BDA0003532769940000072
2) The ratio of the node probabilities that any two nodes with the numbers x and x 'obtain the same output number x' through the mechanism M is obtained
Figure BDA0003532769940000073
Is defined such that it satisfies ∈ -geo-indensinguishability, i.e.
Figure BDA0003532769940000074
First, a proof of 1) is given: let xmax＝(4h+1-1)/3-1，xmin＝(4h-1)/3, then
Figure BDA0003532769940000075
The proof of 2) is given again: according to Pr [ M (x) ═ x']The same node with the number x' is found out, and the constant terms are the same, so that the constant terms are
Figure BDA0003532769940000076
Then is that
Figure BDA0003532769940000077
Cancel each other out. Only the non-constant term is considered below:
Figure BDA0003532769940000078
the inequality of the above formula is obtained by a triangle inequality | a | to | b | less than or equal to | a-b | and the derivation is finished;
step 23, submitting parameters: the user side sends the user type and the disturbance vector
Figure BDA0003532769940000079
And submitting the personalized privacy parameters (e, h) to the server side, and waiting for matching the passenger and the user.
In step 3 of the above method, the distance between the passenger and the driver is confirmed by:
step 331, determining the distance in the same layer: if the positions of the passenger and the driver are in the same layer h of the quartered tree T, the disturbance vectors of the passenger and the driver in the layer h of the quartered tree T are respectively
Figure BDA00035327699400000710
By the following cumulative function
Figure BDA00035327699400000711
Calculating the distance between the two disturbance vectors as the distance between the passenger and the driver; the meaning of each parameter in the accumulation function is as follows:
Figure BDA00035327699400000712
representing a disturbance vector
Figure BDA00035327699400000713
The jth component of (a);
Figure BDA00035327699400000714
representing a disturbance vector
Figure BDA00035327699400000715
The jth component of (a); EMD represents the earth-moving distance of two vectors; in the formula, A: b denotes a is defined as B;
step 332, processing the distance determination of different layers: if the passenger isAt a different level h of the quadtree T than the position of the driver1、h2Floor, passengers and driver h of the quadtree T1、h2The perturbation vectors of the layers are respectively
Figure BDA0003532769940000081
H is to be1、h2After the layers are uniformly polymerized to g layers, g is more than or equal to 0 and less than or equal to H, and disturbance vectors are respectively obtained
Figure BDA0003532769940000082
Perturbation vector aggregated to g-th layer
Figure BDA0003532769940000083
According to the following distance formula
Figure BDA0003532769940000084
Solving the distance between the passenger and the driver at different floors; the meaning of each parameter in the distance formula is as follows:
Figure BDA0003532769940000085
representing a disturbance vector
Figure BDA0003532769940000086
A perturbation vector aggregated to the g-th layer; EMD represents the earth-moving distance of two vectors.
In the above process, h is prepared by1、h2The layers are polymerized to g layers comprising:
(1) when h is larger than or equal to g, adding the value of the node of the previous layer to the father node of the previous layer, and continuously aggregating until the g layer to obtain a one-hot vector;
(2) and when h is less than g, uniformly splitting the value of the current node to four nodes of the next layer, and continuing the splitting until the g layer.
In the above method, the passenger nearest to the driver is determined in the following manner, including:
and sequentially calculating the distance between all passengers and new drivers in the passenger queue at the first floor through a final distance formula, wherein the final distance formula is as follows:
Figure BDA0003532769940000087
in the above formula, pkIndicating passenger queue LpThe kth passenger of (1); h is a total ofkRepresents the height of the tree submitted by the kth passenger; d represents a new driver; h isdIndicating the tree height submitted by the new driver; w (e)k，hk) Is the distance weight of the kth passenger, w (e)k，hk) Is selected as
Figure BDA0003532769940000088
Taking the passenger corresponding to the minimum distance calculated by the final distance formula as the passenger closest to the driver, and taking the passenger from the passenger queue LpDeleting;
determining the driver closest to the passenger in the following manner, comprising:
sequentially calculating driver queues L by a final distance formuladThe distance between all drivers and new passengers on the h-th floor, the final distance formula is:
Figure BDA0003532769940000089
in the above formula, pkShow driver queue LdThe kth driver of (1); h iskIndicating the height of the tree submitted by the kth driver; d represents a new passenger; h isdRepresents the tree height submitted by the new passenger; w (e)k，hk) Distance weight for the kth driver, w (e)k，hk) Is selected as
Figure BDA0003532769940000091
Taking the driver corresponding to the minimum distance calculated by the final distance formula as the driver closest to the passenger, and queuing the driver in the driver queue LdIs deleted.
In the above method, in step 4, the matched passenger and driver establish asymmetric encryption for mutual communication. The safety of communication between the passenger and the driver after matching can be ensured.
In summary, the matching method of the embodiment of the invention provides efficient and effective matching for real-time passengers and drivers on the server side through a personalized position privacy protection mode and two privacy protections with different degrees, thereby protecting position information, providing certain accuracy and ensuring that the two parties are matched according to the closest distance.
In order to more clearly show the technical solutions and the technical effects provided by the present invention, the following detailed description is made on a real-time passenger and driver matching method with personalized location privacy protection provided by the embodiment of the present invention with specific embodiments.
Example 1
As shown in fig. 1 and 2, an embodiment of the present invention provides a method for matching a real-time passenger and a driver in a privacy protection scenario, which mainly includes the following steps: the platform distributes public parameters and models, local personalized noise adding and submitting of user positions, real-time matching of passengers and drivers by the platform, and mutual communication of the passengers and the drivers; wherein,
step 1, the platform distributes public parameters and models: aiming at a two-dimensional plane matrix area R for matching passengers and drivers in real time, a platform (namely a server) establishes a four-branch tree T with the height H as a model corresponding to the two-dimensional plane matrix area R and publishes the model, wherein the model is used for matching the passengers and the drivers in real time: the root node (level 0) of the quadtree T represents the region R, level 1 is four nodes, which are root nodes, i.e. horizontal and vertical uniform quarters of the region R, …, and the node of level H is a horizontal and vertical uniform quarter of the region corresponding to each node of level H-1, up to level H, as shown in fig. 3;
the quartering tree T corresponds to an actual geographic area in the following form, a root node corresponds to an area of a province, four nodes on the 1 st layer represent four areas evenly divided by the province, and so on;
referring to fig. 3, the quad-tree T established as described above has the following properties: 1) level 0 has 1 node, level 1 has 4 nodes, …,the first layer has 4hA node, …, having a level H of 4HEach node corresponds to one region of the region R; 2): each node is numbered from top to bottom, starting from 0 in order from left to right: then the leftmost node of the h-th level is numbered (4)h-1)/3, and the rightmost node is numbered (4)h+1-1)/3-1; 3) the corresponding areas of each layer of nodes are not intersected with each other, and the union of the areas is an area R; 4) each non-leaf node corresponding region is the union of its four child node corresponding regions.
Step 2, local personalized noise adding and submitting of user positions: a user side (including passengers and drivers) takes the region R and the quartered tree T published by the platform, and according to personalized privacy parameters (epsilon, h) of the user side, epsilon represents personalized privacy utility parameters of the user, the value range is [0, + ∞ ], the smaller the epsilon value is, the stronger privacy protection is, the weaker the utility is, the larger the value is, the weaker the privacy protection is, and the stronger the utility is; h represents the height of the user selected to be published on the quadtree T, and the selection range is {0, 1, …, H }, and is also a personalized parameter; if the user is located at a certain position of the plane region R, then at the position of the quadtree T, as shown in fig. 2, a node corresponding to the region where the user is located is represented by red, and the node has the following properties:
1) each layer of nodes has one and only one area where the user pairs are located;
2) the total number of H +1 nodes of the quadtree is the area where the user is located.
In this step, the following steps are mainly divided (see fig. 4):
step 21, the user obtains a private one-hot vector according to the personalized parameters: the user uses a parameter h to indicate the number of the selected issued quartering tree layers, and the larger the h is, the more accurate the area the user wants to issue is, and the lower the privacy protection is; if the area of the user corresponds to the node nh，iThe value of (1) is taken as 1, the values of other nodes are taken as 0, and then the values of all the nodes in the h layer from left to right form a one-hot vector xhAnd is (0, 0, …, 0, 1, 0, …) wherein 1 is at the i-th position.
Step 22, E-geo-identifying uishability privacy protection: for protecting user-issued location informationI.e. protecting one-hot vector xhMaking the privacy protection of the line in the same manner as the line in the same manner; specifically, one-hot vectors correspond to node numbers on the quadtree one by one, and then privacy protection is carried out on the numbers;
the number of the h-th line has a value range of
Figure BDA0003532769940000101
Satisfying the ∈ -geo-indensinguishability mechanism M guarantees that any two numbers x1，x2The probability of disturbing to the same number x' is e∈·d(x，x′)To confine, i.e.
Figure BDA0003532769940000102
Unlike a two-dimensional continuous location scenario, the numbers are one-dimensional discrete and ordered (in most cases, the closer the number, the closer the location to which the number corresponds); here, a Truncated Geometry Mechanism (TGM) is used to satisfy e-geo-induced uishability, formalized as:
Figure BDA0003532769940000103
the requirement of adding the sentiment to the TGM in the standard form of the epsilon-DP can be met by setting the sentiment in the TGM in the standard form of the epsilon-DP to be 1; after the disturbance numbers are numbered, the number is converted into a corresponding disturbance vector
Figure BDA0003532769940000104
Step 23, submitting parameters: user end submits user type (passenger, driver), disturbance vector
Figure BDA0003532769940000105
Personalized privacy parameters (epsilon, h) are sent to a platform, namely a server side, and the platform waits to be matched;
step 3, matching passengers and drivers in real time by the platform: the platform receives the user type and the disturbance vector sent by the user side in real time
Figure BDA0003532769940000111
And personalized privacy parameters (epsilon, h) continuously maintaining the passenger queue LpAnd driver queue LdThen, matching is performed according to the existing user set and the new user, and the following four cases are divided:
(31) the existing user types are only passengers, and the new user types are also passengers: joining newly arrived users to passenger queue Lp；
(32) The existing user type is only the driver, and the new user type is also the driver: joining newly arrived users into driver queue Ld；
(33) The existing user types are passenger only, and the new user types are drivers: queue L of passengerspMatches all passengers and drivers, selects the nearest passenger (i.e. the passenger closest to the new driver), matches the passenger and driver;
defining the distance between the passenger and the driver, determining the closest passenger, comprises:
step 331, distance between the same layers: if the disturbance vectors of the passenger and the driver at the h layer of the four-branch tree are respectively
Figure BDA0003532769940000112
Because both vectors are one-hot vectors, the distance between the two 1's is calculated as their distance (note that in most cases the closer the numbers are, the more adjacent the node corresponding region); the Earth Moving Distance (EMD) just corresponds to the form, and the distance is calculated by adopting an accumulative function to define
Figure BDA0003532769940000113
Step 332, distance between different layers: if the passenger and the driver are not at the same level of the quadtree, at h1，h2Layers, corresponding to disturbance vectors are respectively
Figure BDA0003532769940000114
Because of the different layersThe distances therebetween are not directly comparable, they are polymerized to the g-th layer (g satisfies 0. ltoreq. g. ltoreq.H); the polymerization was divided into two cases according to the values of h and g:
1) when h is larger than or equal to g, adding the value of the current layer node to the father node of the previous layer, and continuously aggregating until the g layer is still a one-hot vector;
2) when h is less than g, the current node value is a father node of four nodes and is an aggregation of the four nodes, and the aggregation mode is unknown, wherein the four nodes are uniformly split and are continuously split until the g level.
Setting disturbance vector
Figure BDA0003532769940000115
The disturbance vector polymerized to the g (0. ltoreq. g.ltoreq.H) th layer is
Figure BDA0003532769940000116
This allows the distance between different layers to be solved (which can be any g)
Figure BDA0003532769940000117
Wherein g is more than or equal to 0 and less than or equal to H;
on the basis of finding the distance, the nearest passenger is determined as follows: now there is a passenger queue LdAnd a new driver, randomly selecting a layer h (h e {0, 1.,. h)mIn which h ismMinimum value representing the current submission tree height of all users), which in turn solves for the passenger queue LpAll passengers and the new driver are at the distance of the h floor
Figure BDA0003532769940000118
Wherein p iskThe k-th passenger, h, of the passenger queuekIndicating the height of the tree submitted by the kth passenger, d indicating the new driver, hdIndicating the tree height submitted by the new driver; considering that the individual parameters of different passengers are different (because the same driver can ignore the influence of the individual parameters), the confidence of the obtained distance is also different, and the passenger weight w (e, h) is added to the distance to balance the influence of the confidence, that is, the final distance is
Figure BDA0003532769940000121
The passenger corresponding to the minimum distance is the passenger closest to the new driver, and the passenger is output and output from the passenger queue LpTo delete the passenger;
(34) the existing user types are only drivers, and the new user types are passengers: the processing is the same as that of the third case (33), and the driver closest to the new passenger can be determined by merely exchanging the calculation of the driver and the passenger.
It can be appreciated that there is no situation where there are both passengers and drivers in the existing user types because of the presence of both, the present invention must take out the passengers or drivers to match until one or both are zero.
Step 4, the passenger and the driver communicate with each other: after the platform matches a pair of passenger and driver, the platform only knows the ID and submitted information of both parties, and does not know any more information. Therefore, the platform tells respective IDs of the two parties, communication is established for the two parties, the two parties can communicate with each other through asymmetric encryption to obtain respective accurate position information in order to ensure that respective information is only known by the other party and no other person knows, and matching is completed.
In summary, the matching method of the embodiment of the invention adopts the quartering tree and truncation geometry mechanism, so that two privacy protections, namely position fuzziness at different levels and differential privacy at different degrees, are possible; meanwhile, due to the adoption of the distance and the matching mode among different layers and the use of asymmetric encryption communication, the matching among users is effective and efficient, the position information can be protected, and certain matching accuracy can be provided.
The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present invention are included in the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims. The information disclosed in this background section is only for enhancement of understanding of the general background of the invention and should not be taken as an acknowledgement or any form of suggestion that this information forms the prior art already known to a person skilled in the art.

Claims (6)

1. A real-time passenger and driver matching method with personalized position privacy protection is characterized in that the method is used in a taxi taking system with a service end in communication connection with a plurality of user ends, and comprises the following steps:
step 1, establishing a model of a real-time matching area: the service end models a two-dimensional plane matrix region R for real-time matching of passengers and drivers into a four-branch tree T with the tree height of H; 1 node on the 0 th layer of the quadtree T is a root node, and the root node represents a region R; the layer 1 is provided with 4 nodes, and each node represents a horizontal, vertical and uniform four equal areas of the area R; the h layer has 4hEach node represents the horizontal, vertical and uniform four equal areas of the corresponding area of each node of the h-1 layer; the H-th layer has 4HEach node represents the horizontal, vertical and uniform four equal areas of the corresponding area of each node of the H-1 layer; numbering each node of the four-way tree T from top to bottom and from left to right from 0 in sequence, wherein the leftmost node of the h-th layer is numbered as (4)h-1) ÷ 3, the node number on the far right being (4)h+1-1)÷3-1；
The server side issues the two-dimensional plane matrix region R and the quadtree T corresponding to the two-dimensional plane matrix region R to all user sides, and the user sides include: passenger and driver clients;
step 2, local personalized noise adding and submitting of the user position: after the user side obtains the two-dimensional plane matrix area R and the quartering tree T corresponding to the two-dimensional plane matrix area R, a private one-hot vector x is obtained according to the personalized privacy parameters (e, h) of the user sidehUsing one-hot vector xhGeneration of perturbed one-hot vectors by truncation of geometric mechanisms
Figure FDA0003532769930000011
Make the one-hot vector
Figure FDA0003532769930000012
The privacy protection of the E-geo-exiting-leaving of the line is met, and the user type and the disturbance vector are transmitted to the user side
Figure FDA0003532769930000013
Submitting the personalized privacy parameters (e, h) to the server;
in the personalized privacy parameters (epsilon, h), epsilon represents the personalized privacy budget parameters of the user terminal, and the value range is [0, + ∞ ]; h represents the height of the user selected to be published on the quadtree T through a user side, and the selection range is {0, 1, …, H };
the user types include: passengers, drivers;
step 3, matching passengers and drivers in real time: the server receives the user type and the disturbance vector sent by the user side in real time
Figure FDA0003532769930000014
And personalized privacy parameters (e, h) and continuously maintaining passenger queues LpAnd driver queue Ld(ii) a Then, the passenger and the driver are matched according to the existing user set and the new user, and the matching steps are as follows:
(31) if the type of the existing user only has passengers and the type of the newly arrived user is also the passenger, adding the newly arrived user into the passenger queue Lp；
(32) If the existing user type is driver only and the new user type is also driver, adding the new user into the driver queue Ld；
(33) If the type of the existing user only has passengers and the type of the new user is a driver, the passenger queue L is arrangedpAll the passengers and the drivers are matched, and the passenger closest to the driver is selected to be matched with the driver;
(34) if the existing user type only has drivers and the new user type is passengers, the drivers are queued to LdAll drivers and passengers are matched, and the driver closest to the passenger is selected to be matched with the passenger;
and 4, establishing communication between the passenger and the driver: and after the server side is matched with a pair of passenger and driver, the respective IDs of the matched passenger and driver are sent to the other side, so that the matched passenger and driver establish communication, namely the matching is completed.
2. The method for matching passengers and drivers with personalized location privacy protection as claimed in claim 1, wherein in the step 2, the user end obtains the private one-hot vector x according to the personalized privacy parameters (e, h) of the user end itselfhUsing one-hot vector xhGeneration of perturbed one-hot vectors by truncation of geometric mechanisms
Figure FDA0003532769930000021
Make the perturbed one-hot vector
Figure FDA0003532769930000022
Obtaining a disturbance vector according to the E-geo-indexing uishability privacy protection processing of the line where the line is positioned
Figure FDA0003532769930000023
The user side sends the user type and the disturbance vector
Figure FDA0003532769930000024
Submitting personalized privacy parameters (epsilon, h) to the server, wherein the parameters comprise:
step 21, the user side obtains a private one-hot vector x according to the personalized parametersh: the user side determines a node n on the quartering tree T corresponding to the area where the user position is located by using a parameter h of the personalized privacy parameter (epsilon, h)h，iAnd connecting the node nh，iIs taken to be 1, the node nh，iThe value of the same-layer node is taken as 0, and the node n is usedh，iAll the nodes of the layer form a one-hot vector x by the values from left to righthThe node nh，iIn the one-hot vector xhAt position i;
step 22, generating a disturbed one-hot vector by using a truncation geometry mechanism
Figure FDA0003532769930000025
Let the one-hot vector
Figure FDA0003532769930000026
Processing satisfying the privacy protection of the e-geo-identifying uishability: one-hot vector xhCorresponding to the number of the node on the T of the quadtree one by one according to the formula
Figure FDA0003532769930000027
Disturbing the two numbers to ensure that the disturbed same number x' meets the privacy protection mechanism M belonging to the E-geo-indentifying uishablility, wherein x in the formula1、x2The numbers of any two nodes on the quadtree T are shown, and x' is the same number formed by disturbance; pr [ M (x)1)＝x′]The representation mechanism M is x at the input1The probability that the time output is x'; d (x, x ') represents the difference between the absolute values of the nodes numbered x, x', i.e., d (x, x ') ═ x-x' |;
disturbing nodes with the number x on the h height in the quadtree T into the number x' according to the following formula, and performing truncation geometric mechanism processing to meet the requirement of E-geo-individissability, wherein the formula is as follows:
Figure FDA0003532769930000028
the meaning of each parameter of the above formula is: x, x' represent the node number,
Figure FDA0003532769930000029
Pr[M(x1)＝x′]the representation mechanism M is x at the input1The probability that the time output is x'; e is a natural constant; h represents the height of the node number x on the quadtree T;
converting the obtained perturbation number x' into correspondingPerturbation vector of
Figure FDA0003532769930000031
Step 23, submitting parameters: the user terminal sends the user type and the disturbance vector
Figure FDA0003532769930000032
And submitting the personalized privacy parameters (e, h) to the server side, and waiting for matching the passenger and the user.
3. The real-time passenger and driver matching method with personalized location privacy protection according to claim 1 or 2, characterized in that in step 3, the distance between the passenger and the driver is confirmed by the following means, including:
step 331, determining the distance in the same layer: if the positions of the passenger and the driver are in the same layer h of the quartered tree T, the disturbance vectors of the passenger and the driver in the layer h of the quartered tree T are respectively
Figure FDA0003532769930000033
By the following cumulative function
Figure FDA0003532769930000034
Calculating the distance between the two disturbance vectors as the distance between the passenger and the driver; the meaning of each parameter in the accumulation function is as follows:
Figure FDA0003532769930000035
representing a disturbance vector
Figure FDA0003532769930000036
The jth component of (a);
Figure FDA0003532769930000037
representing a disturbance vector
Figure FDA0003532769930000038
The jth component of (a); EMD represents the earth-moving distance of two vectors; a in the formula: b denotes a is defined as B;
step 332, processing the distance determination of different layers: if the positions of the passenger and the driver are in different floors h of the quadtree T1、h2Floor, passengers and driver h of the quadtree T1、h2The perturbation vectors of the layers are respectively
Figure FDA0003532769930000039
H is to be1、h2After the layers are uniformly polymerized to g layers, g is more than or equal to 0 and less than or equal to H, and disturbance vectors are respectively obtained
Figure FDA00035327699300000310
Perturbation vector aggregated to the g-th layer
Figure FDA00035327699300000311
According to the following distance formula
Figure FDA00035327699300000312
Solving the distance between the passenger and the driver at different floors; the meaning of each parameter in the distance formula is as follows:
Figure FDA00035327699300000313
representing a perturbation vector
Figure FDA00035327699300000314
A perturbation vector aggregated to the g-th layer; EMD represents the earth-moving distance of two vectors.
4. The method of claim 3, wherein h is selected from the group consisting of1、h2Layer homo-polymerization to g-layer comprising:
(1) when h is larger than or equal to g, adding the value of the previous layer node to the parent node of the previous layer, and continuously aggregating until the g layer to obtain a one-hot vector;
(2) and when h is less than g, uniformly splitting the value of the current node to four nodes of the next layer, and continuing splitting until the g layer.
5. The method of claim 3, wherein determining the passenger closest to the driver comprises:
sequentially calculating passenger queues L through final distance formulapThe distance between all passengers and the new driver on the h-th floor, the final distance formula is:
Figure FDA00035327699300000315
in the above formula, pkIndicating passenger queue LpThe kth passenger of (1); h iskRepresents the height of the tree submitted by the kth passenger; d represents a new driver; h isdIndicating the tree height submitted by the new driver; w (e)k，hk) Is the distance weight of the kth passenger, w (e)k，hk) Is selected as
Figure FDA0003532769930000041
Taking the passenger corresponding to the minimum distance calculated by the final distance formula as the passenger closest to the driver, and taking the passenger from the passenger queue LpDeleting;
determining the driver closest to the passenger in the following manner, comprising:
sequentially calculating driver queues L by a final distance formuladThe distance between all drivers and new passengers on the h-th floor, the final distance formula is:
Figure FDA0003532769930000042
in the above formula, pkShow driver queue LdTo (1) ak drivers; h iskIndicating the height of the tree submitted by the kth driver; d represents a new passenger; h isdRepresents the tree height submitted by the new passenger; w (e)k，hk) Distance weight for the kth driver, w (e)k，hk) Is selected as
Figure FDA0003532769930000043
Taking the driver corresponding to the minimum distance calculated by the final distance formula as the driver closest to the passenger, and queuing the driver in the driver queue LdIs deleted.
6. The method for matching passengers and drivers with personalized location privacy protection as claimed in claim 1 or 2, wherein in step 4, the matched passengers and drivers establish mutual communication by asymmetric encryption.
CN202210212477.0A 2022-03-04 2022-03-04 Real-time passenger and driver matching method with personalized location privacy protection Active CN114595480B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210212477.0A CN114595480B (en) 2022-03-04 2022-03-04 Real-time passenger and driver matching method with personalized location privacy protection

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210212477.0A CN114595480B (en) 2022-03-04 2022-03-04 Real-time passenger and driver matching method with personalized location privacy protection

Publications (2)

Publication Number Publication Date
CN114595480A true CN114595480A (en) 2022-06-07
CN114595480B CN114595480B (en) 2024-04-02

Family

ID=81815610

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210212477.0A Active CN114595480B (en) 2022-03-04 2022-03-04 Real-time passenger and driver matching method with personalized location privacy protection

Country Status (1)

Country Link
CN (1) CN114595480B (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109302676A (en) * 2018-10-17 2019-02-01 长安大学 A kind of O-D Region Matching method retained towards passenger and driver's privacy
CN109544900A (en) * 2018-11-21 2019-03-29 长安大学 A kind of route matching method that the privacy multiplying trip altogether towards passenger and driver retains
WO2020199524A1 (en) * 2019-04-02 2020-10-08 长安大学 Method for matching ride-sharing travellers based on network representation learning
CN112752232A (en) * 2021-01-07 2021-05-04 重庆大学 Privacy-oriented driver-passenger matching mechanism
CN114118618A (en) * 2021-12-06 2022-03-01 中国人民解放军国防科技大学 Passenger matching method, device, equipment and storage medium for privacy protection car pooling

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109302676A (en) * 2018-10-17 2019-02-01 长安大学 A kind of O-D Region Matching method retained towards passenger and driver's privacy
CN109544900A (en) * 2018-11-21 2019-03-29 长安大学 A kind of route matching method that the privacy multiplying trip altogether towards passenger and driver retains
WO2020199524A1 (en) * 2019-04-02 2020-10-08 长安大学 Method for matching ride-sharing travellers based on network representation learning
CN112752232A (en) * 2021-01-07 2021-05-04 重庆大学 Privacy-oriented driver-passenger matching mechanism
CN114118618A (en) * 2021-12-06 2022-03-01 中国人民解放军国防科技大学 Passenger matching method, device, equipment and storage medium for privacy protection car pooling

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
蔡威;白光伟;沈航;成昭炜;张慧丽;: "移动群智感知中基于强化学习的双赢博弈", 计算机科学, no. 10, 15 October 2020 (2020-10-15) *

Also Published As

Publication number Publication date
CN114595480B (en) 2024-04-02

Similar Documents

Publication Publication Date Title
Wimsatt False models as means to truer theories
CN103577888B (en) Method for optimization of product design schemes
US20120078957A1 (en) Graph Server Querying for Managing Social Network Information Flow
CN111291270B (en) Attribute reasoning and product recommendation method based on self-adaptive graph convolution network
CN109741120B (en) Product deep personalized customization method for user requirements
CN102750647A (en) Merchant recommendation method based on transaction network
CN107766742B (en) Multi-correlation differential privacy matrix decomposition method under non-independent same-distribution environment
CN109615452A (en) A kind of Products Show method based on matrix decomposition
Abel et al. Group aggregation of pairwise comparisons using multi-objective optimization
CN114595480A (en) Real-time passenger and driver matching method with personalized location privacy protection
US20230065902A1 (en) Intention-driven interactive form-filling method for dikw cotent
Rakhmawati et al. Optimal path with interval value of intuitionistic fuzzy number in multigraph
Ping et al. New approach for quality function deployment with an extended alternative queuing method under linguistic Pythagorean fuzzy environment
Meneghini et al. Information to the eye of the beholder: Data visualization for many-objective optimization
CN114218337A (en) Natural resource survey monitoring data identification and fusion updating method
Zhou et al. K-medoids method based on divergence for uncertain data clustering
CN108959956A (en) Difference private data dissemination method based on Bayesian network
Izadikhah Deriving weights of criteria from inconsistent fuzzy comparison matrices by using the nearest weighted interval approximation
CN107908630A (en) Material picture color classification retrieving method
WO2023029189A1 (en) Dikw content-oriented intention-driven interactive form filling method
CN113536109B (en) Interest point recommendation method based on neural network and mobile context
CN104679988A (en) TOPSIS (techniques for order preference by similarity to ideal solution)-based multi-attribute decision-making method
CN112612207B (en) Multi-target game solving method and system under uncertain environment
CN113849772A (en) Matrix decomposition recommendation method based on localized differential privacy
Thangaraj et al. Authorship Pattern on Multiple Disabilities during 2006-2020: A Scientometric Analysis

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant